Life Insurance and Annuities: The Impacts of Regulatory Requirements on Consumer Cost and Consumer Choice
|
 For a print-ready PDF of this page, click here.
|
|||||||||
What This Issue Brief Does Not Do
This issue brief does not attempt to address the value to the public of any mandates proposed, enacted, or implemented. The purpose of this issue brief is to inform policymakers about the impact that various types of mandates can have on consumer cost and consumer choice for annuity and life insurance products.
Equality and Equity
Some mandates that conflict with these principles have a greater impact than do others. Some have more far-reaching consequences than do others. Understanding the impact of a conflict with these principles helps one to evaluate the long-term effect of a proposed mandate on consumer cost and consumer choice.
In an open market, consumers are free to choose which products to buy and how much to buy. Their decisions to buy are based mostly on their own view of their needs and of the value of the products to them.
Basic insurance principles require that, in a largely open market, the price charged (the “premium”) for the insurance be based on cost. This means assessing the expected cost of the benefits for all consumers buying the product based on equity; the higher the expected cost, the higher the premium should be. Note that expected benefit costs account for both the amount and the timing of the benefit payments.
Equity demands that premium rates be comparable for insurance risks with similar expected benefit costs—as justified by credible data—and be different for risks with different expected costs. This induces the insurer to classify risks by expected costs.
An equitable price might not be the same as an equal price. The following simple, hypothetical cases involving people in two age groups can illustrate. The first three cases, which do not represent an open market, are successively modified to create additional cases that will illustrate the basic principles central to this issue brief.
Case 1. Suppose we have a population of 20,000 people. Half are age 25 and half are age 65. Everyone is required to buy a $100,000 life insurance policy from XYZ Insurance Company. Another law also requires that the same premium be charged for each policy. This is not an open market because consumers are not free to choose whether to buy or how much to buy.
XYZ has credible data for these people. In this case, XYZ can easily determine total expected benefit costs; it knows who will be insured and for how much. It can easily calculate a uniform rate of premium to be charged for each policy.
Case 2. Suppose the population comprises only 10 people age 25 and 10 people age 65, instead of 10,000 at each age. Because the laws of Case 1 still apply, this also is not an open market.
In this case, XYZ would be less sure of the total expected benefit costs because the net effect of the timing of the benefit payments would be less predictable. XYZ would add a “small-pool risk margin” to the premium charged, to protect from the risk of a larger statistical variance in the timing of the deaths.
Case 3. Back to a population of 20,000 people, suppose that the mix of people by age is slightly different—12,000 at age 25 and 8,000 at age 65.
XYZ still could easily calculate a uniform rate of premium to be charged for each policy. The rate would be lower than in the first case because the total expected cost of the benefits would be lower. This structure also can be financially sound because everyone must buy a policy from XYZ and for the same amount. As in the first two cases, this is not an open market.
Each of the following life insurance cases represents a market where consumers have choices.
In this case, unlike Case 4, the people age 25 will not be subsidizing higher-cost consumers for their policies. They are more likely to view the policy as a good value to them.
Case 6. This case is the same as Case 5, with one addition. A group of 10,000 people age 80 has been added to the population of Case 5.
XYZ does not have sufficient credible data for people that age. XYZ knows that there is a risk in making a bad guess about the expected benefit costs for these people. XYZ will likely add a risk margin for insufficient data to the premium for people age 80. This added margin increases the premium rate for every person age 80 who buys a policy. It will exceed the rate based on the expected costs using only the insufficient data.
The simple cases above illustrate the five basic insurance principles that apply in an open market:
Let’s assume that XYZ Insurance Company offers a lifetime annuity. It pays income of $1,000 each month for the rest of the person’s life.
Case 7. This case has the same population as in Case 1 above—10,000 people age 25 and 10,000 people age 65. The law in this case requires everyone to buy a $1,000 lifetime annuity from XYZ Insurance Company. The law also requires that the same one-time (or “single”) premium is charged for each annuity. As in Case 1, this is not an open market because consumers are not free to choose whether to buy or how much to buy.
XYZ has credible data for these people. The expected benefit costs are much higher for a person age 25 than for a person age 65. XYZ can calculate a uniform rate of single premium to cover total expected costs for all annuities bought. It would be an amount between the (higher) single-premium rate for a person age 25 and the (lower) rate for a person age 65.
Case 8. This case is the same as Case 7, with one change. The requirement to buy an annuity is removed. Each person can freely choose whether to buy one from XYZ. But the single-premium rates still must be equal for all annuities.
In this case, as in Case 4, XYZ must estimate the mix of sales by age in order to set a uniform single-premium rate that will cover the total expected benefit costs for all annuities bought.
Also similar to Case 4 but in reverse, people age 65 are less likely to buy an annuity than people age 25; they would view it as a poor value to them. Many would believe that they’re paying extra to subsidize the people age 25, and they can choose not to do it.
On the other hand, the people age 25 would view it as a good value to them. They would feel that they’re getting a bargain. They would choose to buy at this “bargain” price and select against the insurer. Antiselection would occur.
So XYZ likely will add a mix-of-sales risk margin to the uniform single-premium rate that otherwise is based exactly on the mix of sales it expects. This added margin raises the cost for each person who buys an annuity. Also as in Case 4, a rate spiral can occur.
Case 9. This case is the same as Case 8, with one change. The single-premium rates need not be equal for all annuities.
In this case, as in Case 5, XYZ is able to set single-premium rates based on the age of the person. The rates for people age 25 and for people age 65 would reflect the expected benefit costs for that age. XYZ would not need to add a mix-of-sales risk margin to each single premium.
To set the single-premium rate for each annuity bought, XYZ would underwrite each person, considering only the age. This allows XYZ to equitably classify each risk by age and set the single premium. And in this way, the person choosing to buy cannot select against the insurer. The single premium charged for each annuity equitably reflects the expected benefit cost for that annuity.
In this case, unlike Case 8, the people age 65 will not be subsidizing higher-cost consumers for their annuities. They more likely will view the annuity as a good value to them.
These simple, hypothetical annuity and life insurance cases will be referred to in the remainder of this issue brief in describing the kinds of mandates that can impact consumer cost and consumer choice.
Mandates That Limit Insurers’ Ability to Equitably Classify Risks on New Policies
To remain viable, an insurer needs to charge premiums at rates sufficient to cover expected benefit costs.
A setting like in Case 1 or Case 3—where the insurer knows the nature of the risk in advance and has high confidence in its estimate of expected benefit costs—does not exist in an open market. In an open market, the setting is usually similar to the one in Case 5. Each person who chooses to buy is underwritten. The insurer classifies the risk and then charges a price based on its expected benefit cost from that assessment.
For more about how insurers classify risks, or “risk classification,” see On Risk Classification,1 the Academy Risk Classification Work Group’s November 2011 monograph on the subject.
The less certain an insurer is about the expected cost of benefits, the higher the total risk margin the insurer is inclined to add to the premium. This margin can include a small-pool risk margin (described in Case 2) and a mix-of-sales risk margin (Case 4). Often, all consumers who buy from the insurer pay these risk margins, whether they individually contribute to the risk or not.
The insufficient-data risk margin (first described in Case 6) reflects uncertainty about the expected benefit cost of a risk due to an insufficient amount of credible data about that risk. For example, uncertainty about the amount of extra mortality risk linked to a certain high-risk activity, such as hang gliding, might cause insurers to add an insufficient-data risk margin to the premium rate only for a person expecting to take part in the activity—or instead (in some cases, at the person’s option) to exclude benefit coverage for death from such high-risk activity. This approach of impacting only those expecting to participate in high-risk activities allows insurers to keep their rates lower for consumers who do not expect to participate in such high-risk activities. For an example of a mandate to which this applies, see the section below on foreign travel.
Situations like the one in Case 4—where a mandate allows people of different ages to choose whether to buy but requires that the price be the same for everyone—are not equitable to consumers:
In these cases, the mandates can cause some consumers to pay a higher price for their coverage even though they are not higher-cost risks, and can cause other consumers to choose to go without coverage that they may need rather than pay the higher price. Also, because fewer policies are sold, the insurer has to consider the increased small-pool risk (Case 2), which could further increase the premiums paid by all consumers who buy coverage.
In some cases, an insurer might conclude that a particular mandate that limits insurers’ ability to equitably classify risks creates a market environment in which the insurer cannot be successful. In such cases, the insurer may choose not to participate in the market. This limits consumer choices in the market.
In the best cases, when an insurer expects little variance in benefit costs and little or no antiselection, then the total risk margin added to premium rates is lower and consumers pay a lower cost for the benefit.
Listed below are some mandates, enacted or considered, that limit insurers’ ability to equitably classify risks and charge for them, and what the results have been or may be.
HIV/AIDS—When the HIV/AIDS crisis emerged, life (and health) insurers would not insure people who had tested positive for, were being treated for, or showed symptoms of HIV/AIDS. Expected benefit costs were either too high or too uncertain. Some insurers considered HIV/AIDS testing for all people applying for policies. But such a practice would increase costs for all consumers who buy policies.
Policymakers wanted all people to have an “equal right” to buy life insurance without having to share information about their HIV/AIDS illness. Insurers maintained that the issue for them wasn’t privacy or unfair discrimination but rather cost. The expected cost of the antiselection risk was too high for all consumers to pay. A rate spiral would occur. This was illustrated above in Case 4 (albeit based on age, not medical condition) and in the fourth basic insurance principle. It is deemed no different than mandating the same rule for other people who know they have a serious medical condition, such as advanced-stage cancer.
Note that many employer-sponsored pension plans are subject to a similar gender-neutral limit. But these plans are really more like Case 7, where there is a closed system and people do not choose whether to participate or how much to buy. It is not an open market, and it can be structured to be financially sound for that reason.
Mandates That Require Certain Rights or Benefits on New Policies
Like many other products offered in an open market, life insurance policies and annuity contracts can be designed with many “built-in” features and benefits. For example, some life policies include a built-in accidental death benefit, whereas others do not—just as some automobiles come with climate control as “standard equipment,” and, with others, it is offered as an extra-cost option.
Whether it is life insurance or new cars, a company designs a product with the combination of features it thinks customers will value the most. Why is this important? In an open market, consumers will decide to purchase based on their views about the value of the product to them.
When new (or increased) contractual rights or benefits for annuity or life insurance products are required by mandate, the expected extra costs of the mandate can be large or small. If they are perceived by an insurer to be significant, the insurer would generally have the following two options for recovering the extra cost of the benefit added to new policies:
Listed below are some examples of mandates that increase the cost of insurance for all consumers. The first example below is actually a case of a long-standing mandate becoming a significant cost issue for consumers due to an abrupt change in economic conditions.
Fixed policy-loan rates—Life insurance policies with cash values normally provide for loans. To set policy rates, insurers must estimate how much of the invested assets supporting policy values normally held in the company’s general investment account will be diverted to policy loans. If the loan interest rate is much lower than the rate earned on other investments, the impact can be large. When more is invested in policy loans at a relatively low rate, less can be invested elsewhere at a relatively higher rate. This impacts the overall investment returns that can be allocated among all policies to keep down costs.
Historically, state insurance laws have mandated that interest can be charged for policy loans at a fixed annual rate not to exceed some maximum, such as 6 or 8 percent. During the 1980s, rates of return available on personal investments spiked. Consumers began to realize that they could borrow from their policies at a relatively low rate of interest and invest that borrowed money in the market for an easy profit. In fact, some consumers were purchasing new policies just for that purpose. This is a form of interest-rate antiselection called “disintermediation.”
When insurers had to sell off long-term investments in order to obtain the liquid capital needed to provide these policy loans, the losses realized on those investments were either borne by investors (stock insurers) or passed along to policyowners in the form of lower policy dividends (mutual insurers), raising policy costs for all consumers.
To address this antiselection problem, insurers developed an “adjustable loan rate” provision. Until state laws were changed to allow insurers to use this new provision, all consumers who had policies with fixed loan rates were exposed to the risk of this antiselection.
‘Free-look’ provision—All life insurance policies and annuity contracts have a right-to-return, or “free-look,” provision. It provides the new contract owner a period of time after receiving the contract to review it and, if not satisfied with it, the right to return it for a full refund without penalty. State insurance laws generally require that this free-look period run for at least 10 days after taking delivery of the new contract.
The mandated right to return a newly entered contract—one involving a significant financial commitment and containing several provisions that might not be thoroughly discussed and understood at the time of application—is not unique to the life insurance industry. But unlike other types of contracts, the life insurance coverage during the free-look period is not “free.” These contracts do not delay commencement of coverage until completion of the free-look period.
Deaths do occur before the end of the free-look period, and death claims are paid for deaths during this period. Hence, there is a real cost for the life insurance coverage provided during this period. And the cost for this coverage under policies returned under the free-look period is borne by all the other contract owners—consumers of insurance coverage—who do not return their policies under this provision.
To extend the free-look period only increases the costs borne by the remaining consumers who purchased coverage. This issue can be exacerbated in cases where the insurer also is limited by another mandate that limits risk classification, such as not being able to consider the foreign-travel plans of applicants.
For example, with a mandated 30-day free-look period and such a mandated limit on risk classification, a person planning a trip to a potentially dangerous region might look at this combination as a way to obtain a complete medical examination (through the underwriting process) and a high amount of valuable “travel insurance” for free—paid for by the other consumers who purchase insurance coverage with the insurer and don’t return it. As with the fixed policy-loan rates, this can be a form of anti selection.
So, to extend the currently mandated free-look period could increase consumer costs—an unintended result—in several possible ways.
Surrender charges on flexible-premium contracts—State nonforfeiture laws mandate that owners of annuity and life insurance contracts who surrender their contracts do not forfeit the unused value of their accumulated premium payments. For fixed-premium life insurance contracts such as whole life, the mandate prescribes minimum cash surrender values calculated assuming recovery of insurers’ initial sales and administrative expenses is amortized over time.
Flexible-premium contracts such as universal life accumulate account values, based on actual premium amounts paid, that will vary from contract to contract. Deducting a surrender charge from the account value on surrender is a way to recover such unamortized initial expenses at that time. So nonforfeiture laws for flexible-premium contracts attempt to limit surrender charges on these contracts in a way intended to generate cash surrender values on a par with those for fixed-premium contracts.
To further limit surrender charges on flexible-premium contracts can be viewed by policymakers as increasing benefits to consumers who surrender their contracts for cash. While this might be true for surrendering contract owners, it would increase the losses that must be passed along to the remaining contract owners in higher costs. In effect, the remaining contract owners would be paying the cost of the additional cash benefits provided to the surrendering contract owners—an unintended result of such a mandate.
Minimum death benefits relative to premiums—Life insurance protection is needed by many types of consumers for many purposes. In some cases, it can be used by older consumers as a way to defray the costs of final expenses. The benefit amounts needed in these cases usually are relatively small. To meet the needs of the final-expenses market, classification (underwriting) requirements are often relaxed, which increases the expected average costs per thousand dollars of benefit. Because of this, the higher average age of buyers, and the fixed business expenses that must be covered by all policies, the premiums for these lower-amount policies relative to the benefits normally are higher than for most other policies.
Several years ago, policymakers in one jurisdiction mandated that, for most life insurance policies with such smaller benefit amounts, the death benefit cannot be lower than the premiums, accumulated at interest, during the first several years.2 This is a special case in this category of mandating benefits—it’s really mandating that premiums meet certain limits.
Similar to the result mentioned above in the HIV/AIDS section, several life insurers unilaterally decided not to offer lower-amount life insurance policies in this jurisdiction. And similar to the HIV/AIDS situation, this meant that all consumers in this jurisdiction needing lower amounts had fewer product choices.
This mandate has existed for many years. It originally applied to all life insurance policies with a minimum death benefit under $25,000 and was based on accumulating premiums at a fixed rate of 5 percent. In recent years, however, the mandate has been scaled back to affect only policies under $5,000 and accumulate premiums at a lower rate that is based on current market rates. Now, many more consumers in this market have more product choices.
Equality and Equity
Mandates for annuity and life insurance benefits are often intended to either require benefits that are fair to consumers or protect consumers from unfair treatment. What policymakers view as fair or unfair may be based on their view of equality. A person’s view of the meaning of “equality” can vary from situation to situation. In the context of insurance benefits, people generally view it to mean that consumers who purchase the same level of benefits should pay the same level of premiums.
In a completely open market, anyone can choose whether to buy or sell a product, how much to buy or sell, and at what price to buy or sell. Insurance companies currently sell their annuity and life insurance products in a largely open market, somewhat restricted by regulatory mandates.
To succeed in such a largely open market, insurers must price and sell annuity and life insurance products based on the concept of equity, which at times can differ from equality. Equity requires that the value of the premiums reflect the value of the benefits. So, if the value of the benefits for a person age 65 differs from the value for a person age 25, then the value of the premiums payable by each must reflect the differing values of the benefits.
To remain solvent, insurers must classify—and charge for—risks based on expected benefit costs. These expected costs are based on equity.
For example, charging a person age 25 and someone age 65 the same price for life insurance makes the prices equal, but it isn’t fair or equitable because the expected cost of providing the same level of benefit for a person age 65 is greater. Consumers will choose what to buy based on their own views about the value of the product to them. Such views often are based on equity.
The basic insurance principles that apply to annuity and life insurance benefits in an open market are based on equity. Mandates that conflict with these principles can impact consumer cost and consumer choice. The impact can be in the form of:
In a completely open market, anyone can choose whether to buy or sell a product, how much to buy or sell, and at what price to buy or sell. Insurance companies currently sell their annuity and life insurance products in a largely open market, somewhat restricted by regulatory mandates.
To succeed in such a largely open market, insurers must price and sell annuity and life insurance products based on the concept of equity, which at times can differ from equality. Equity requires that the value of the premiums reflect the value of the benefits. So, if the value of the benefits for a person age 65 differs from the value for a person age 25, then the value of the premiums payable by each must reflect the differing values of the benefits.
To remain solvent, insurers must classify—and charge for—risks based on expected benefit costs. These expected costs are based on equity.
For example, charging a person age 25 and someone age 65 the same price for life insurance makes the prices equal, but it isn’t fair or equitable because the expected cost of providing the same level of benefit for a person age 65 is greater. Consumers will choose what to buy based on their own views about the value of the product to them. Such views often are based on equity.
The basic insurance principles that apply to annuity and life insurance benefits in an open market are based on equity. Mandates that conflict with these principles can impact consumer cost and consumer choice. The impact can be in the form of:
- Higher prices for some or all consumers;
- Reduced benefits for some or all consumers; or
- Fewer product choices for all consumers.
Some mandates that conflict with these principles have a greater impact than do others. Some have more far-reaching consequences than do others. Understanding the impact of a conflict with these principles helps one to evaluate the long-term effect of a proposed mandate on consumer cost and consumer choice.
Basic Insurance Principles That Apply
Simply put, insurance protects against financial risk. For example:
- Auto insurance protects against the financial risk of a car accident;
- Life insurance protects against the financial risk of untimely death; and
- A lifetime annuity protects against the financial risk of outliving one’s assets.
In an open market, consumers are free to choose which products to buy and how much to buy. Their decisions to buy are based mostly on their own view of their needs and of the value of the products to them.
Basic insurance principles require that, in a largely open market, the price charged (the “premium”) for the insurance be based on cost. This means assessing the expected cost of the benefits for all consumers buying the product based on equity; the higher the expected cost, the higher the premium should be. Note that expected benefit costs account for both the amount and the timing of the benefit payments.
Equity demands that premium rates be comparable for insurance risks with similar expected benefit costs—as justified by credible data—and be different for risks with different expected costs. This induces the insurer to classify risks by expected costs.
An equitable price might not be the same as an equal price. The following simple, hypothetical cases involving people in two age groups can illustrate. The first three cases, which do not represent an open market, are successively modified to create additional cases that will illustrate the basic principles central to this issue brief.
Case 1. Suppose we have a population of 20,000 people. Half are age 25 and half are age 65. Everyone is required to buy a $100,000 life insurance policy from XYZ Insurance Company. Another law also requires that the same premium be charged for each policy. This is not an open market because consumers are not free to choose whether to buy or how much to buy.
XYZ has credible data for these people. In this case, XYZ can easily determine total expected benefit costs; it knows who will be insured and for how much. It can easily calculate a uniform rate of premium to be charged for each policy.
Case 2. Suppose the population comprises only 10 people age 25 and 10 people age 65, instead of 10,000 at each age. Because the laws of Case 1 still apply, this also is not an open market.
In this case, XYZ would be less sure of the total expected benefit costs because the net effect of the timing of the benefit payments would be less predictable. XYZ would add a “small-pool risk margin” to the premium charged, to protect from the risk of a larger statistical variance in the timing of the deaths.
Case 3. Back to a population of 20,000 people, suppose that the mix of people by age is slightly different—12,000 at age 25 and 8,000 at age 65.
XYZ still could easily calculate a uniform rate of premium to be charged for each policy. The rate would be lower than in the first case because the total expected cost of the benefits would be lower. This structure also can be financially sound because everyone must buy a policy from XYZ and for the same amount. As in the first two cases, this is not an open market.
Each of the following life insurance cases represents a market where consumers have choices.
Case 4. This case is the same as Case 1, with one change. The requirement to buy a policy is removed. Each person can freely choose whether to buy a $100,000 policy from XYZ. But the premiums still must be equal for all policies.
In this case, XYZ must estimate how many people in each age group will choose to buy a policy from XYZ in order to set a uniform premium rate that will cover the total expected benefit costs for all policies bought.
People age 25 would be less likely to buy a policy than people age 65. The younger people would believe that they’re paying extra to subsidize the people age 65, and they would view it as a poor value.
On the other hand, the people age 65 would view it as a good value to them. They would feel that they’re getting a bargain. In other words, they would choose, or select, to buy at this bargain price. In this example, the insurer cannot use knowledge of the person’s age to set the premium. The person choosing to buy knows more about the risk than the insurer knows and can use that knowledge to select against the insurer. This is called “antiselection” in the industry. In the extreme, antiselection can impact the insurer’s solvency.
XYZ knows that there is a risk in making a bad guess about the “mix of sales” by age. So XYZ will likely add a “mix-of-sales risk margin” to the uniform premium that otherwise is based exactly on the mix of sales it expects. This added risk margin raises the premium for each person who chooses to buy a policy.
Even with this margin added to the premium, many people age 65 still would view the policy as a good value to them, and most of them, on that basis, would choose to buy a policy.
If XYZ’s guess is wrong, and the portion of people age 25 buying a policy is lower than even the assumption used to calculate the mix-of-sales risk margin, then XYZ will lose money on that pool of risks. Its only choice for future sales is to raise the uniform premium even more. But so doing will, in turn, lower the value of the policy to people age 25; even fewer of them will buy new policies.
The uniform premium rate will increase each successive year. This is known in the industry as a “rate spiral.” Fewer and fewer people age 25 will choose to buy policies because the ever-increasing rate will become less and less attractive to them. Eventually, only people age 65 will choose to buy a policy because the uniform premium rate will eventually be equal to the rate for people that are all age 65.
Case 5. This case is the same as Case 4, with one change. The premium rates need not be equal for all policies.
In this case, XYZ is able to set premium rates based on the age of the person. The premium rates for people age 25 would reflect the expected benefit costs for that age—likewise for people age 65. XYZ would not need to add a mix-of-sales risk margin to each premium.
To set the premium for each policy bought, XYZ would need to underwrite each person. This simple underwriting process considers only the age of the person. It allows XYZ to equitably classify each risk by age and set the premium. And in this way, the insurer knows what the person choosing to buy knows about the risk; the person choosing to buy cannot select against the insurer. The premium charged for each policy equitably reflects the expected benefit cost for that policy.
In this case, XYZ must estimate how many people in each age group will choose to buy a policy from XYZ in order to set a uniform premium rate that will cover the total expected benefit costs for all policies bought.
People age 25 would be less likely to buy a policy than people age 65. The younger people would believe that they’re paying extra to subsidize the people age 65, and they would view it as a poor value.
On the other hand, the people age 65 would view it as a good value to them. They would feel that they’re getting a bargain. In other words, they would choose, or select, to buy at this bargain price. In this example, the insurer cannot use knowledge of the person’s age to set the premium. The person choosing to buy knows more about the risk than the insurer knows and can use that knowledge to select against the insurer. This is called “antiselection” in the industry. In the extreme, antiselection can impact the insurer’s solvency.
XYZ knows that there is a risk in making a bad guess about the “mix of sales” by age. So XYZ will likely add a “mix-of-sales risk margin” to the uniform premium that otherwise is based exactly on the mix of sales it expects. This added risk margin raises the premium for each person who chooses to buy a policy.
Even with this margin added to the premium, many people age 65 still would view the policy as a good value to them, and most of them, on that basis, would choose to buy a policy.
If XYZ’s guess is wrong, and the portion of people age 25 buying a policy is lower than even the assumption used to calculate the mix-of-sales risk margin, then XYZ will lose money on that pool of risks. Its only choice for future sales is to raise the uniform premium even more. But so doing will, in turn, lower the value of the policy to people age 25; even fewer of them will buy new policies.
The uniform premium rate will increase each successive year. This is known in the industry as a “rate spiral.” Fewer and fewer people age 25 will choose to buy policies because the ever-increasing rate will become less and less attractive to them. Eventually, only people age 65 will choose to buy a policy because the uniform premium rate will eventually be equal to the rate for people that are all age 65.
Case 5. This case is the same as Case 4, with one change. The premium rates need not be equal for all policies.
In this case, XYZ is able to set premium rates based on the age of the person. The premium rates for people age 25 would reflect the expected benefit costs for that age—likewise for people age 65. XYZ would not need to add a mix-of-sales risk margin to each premium.
To set the premium for each policy bought, XYZ would need to underwrite each person. This simple underwriting process considers only the age of the person. It allows XYZ to equitably classify each risk by age and set the premium. And in this way, the insurer knows what the person choosing to buy knows about the risk; the person choosing to buy cannot select against the insurer. The premium charged for each policy equitably reflects the expected benefit cost for that policy.
In this case, unlike Case 4, the people age 25 will not be subsidizing higher-cost consumers for their policies. They are more likely to view the policy as a good value to them.
Case 6. This case is the same as Case 5, with one addition. A group of 10,000 people age 80 has been added to the population of Case 5.
XYZ does not have sufficient credible data for people that age. XYZ knows that there is a risk in making a bad guess about the expected benefit costs for these people. XYZ will likely add a risk margin for insufficient data to the premium for people age 80. This added margin increases the premium rate for every person age 80 who buys a policy. It will exceed the rate based on the expected costs using only the insufficient data.
The simple cases above illustrate the five basic insurance principles that apply in an open market:
- To remain viable, insurers need to charge premiums at rates sufficient to cover expected benefit costs (all cases);
- If insurers are unsure of the level of expected benefit costs, they are likely to add extra risk margins to the rates in order to cover the risk of uncertainty (Case 2, Case 4, and Case 6), which increases consumer costs;
- Consumers will decide to buy a product based on their view about the value of that product to them (Case 4 and Case 5);
- If consumers can choose whether to buy the product and insurers are not permitted to use what consumers know about their own risks in setting rates, then consumers who believe they are high-cost risks will tend to select against the insurer (Case 4); and
- If insurers can underwrite each risk and can equitably classify the risks, then insurers are more certain about the level of expected benefit costs and are less likely to add extra risk margins (Case 5), which will keep consumer costs down.
Let’s assume that XYZ Insurance Company offers a lifetime annuity. It pays income of $1,000 each month for the rest of the person’s life.
Case 7. This case has the same population as in Case 1 above—10,000 people age 25 and 10,000 people age 65. The law in this case requires everyone to buy a $1,000 lifetime annuity from XYZ Insurance Company. The law also requires that the same one-time (or “single”) premium is charged for each annuity. As in Case 1, this is not an open market because consumers are not free to choose whether to buy or how much to buy.
XYZ has credible data for these people. The expected benefit costs are much higher for a person age 25 than for a person age 65. XYZ can calculate a uniform rate of single premium to cover total expected costs for all annuities bought. It would be an amount between the (higher) single-premium rate for a person age 25 and the (lower) rate for a person age 65.
Case 8. This case is the same as Case 7, with one change. The requirement to buy an annuity is removed. Each person can freely choose whether to buy one from XYZ. But the single-premium rates still must be equal for all annuities.
In this case, as in Case 4, XYZ must estimate the mix of sales by age in order to set a uniform single-premium rate that will cover the total expected benefit costs for all annuities bought.
Also similar to Case 4 but in reverse, people age 65 are less likely to buy an annuity than people age 25; they would view it as a poor value to them. Many would believe that they’re paying extra to subsidize the people age 25, and they can choose not to do it.
On the other hand, the people age 25 would view it as a good value to them. They would feel that they’re getting a bargain. They would choose to buy at this “bargain” price and select against the insurer. Antiselection would occur.
So XYZ likely will add a mix-of-sales risk margin to the uniform single-premium rate that otherwise is based exactly on the mix of sales it expects. This added margin raises the cost for each person who buys an annuity. Also as in Case 4, a rate spiral can occur.
Case 9. This case is the same as Case 8, with one change. The single-premium rates need not be equal for all annuities.
In this case, as in Case 5, XYZ is able to set single-premium rates based on the age of the person. The rates for people age 25 and for people age 65 would reflect the expected benefit costs for that age. XYZ would not need to add a mix-of-sales risk margin to each single premium.
To set the single-premium rate for each annuity bought, XYZ would underwrite each person, considering only the age. This allows XYZ to equitably classify each risk by age and set the single premium. And in this way, the person choosing to buy cannot select against the insurer. The single premium charged for each annuity equitably reflects the expected benefit cost for that annuity.
In this case, unlike Case 8, the people age 65 will not be subsidizing higher-cost consumers for their annuities. They more likely will view the annuity as a good value to them.
These simple, hypothetical annuity and life insurance cases will be referred to in the remainder of this issue brief in describing the kinds of mandates that can impact consumer cost and consumer choice.
Potential Impact of Mandates on Consumers
Policymakers propose mandates intended for the common good. But sometimes these mandates can have unintended results. The kinds of mandates we consider here:
- Limit insurers’ ability to equitably classify risks on new policies; and
- Require certain rights or benefits on new policies.
Each of these kinds of mandates can cause unintended results for consumers.
Mandates That Limit Insurers’ Ability to Equitably Classify Risks on New Policies
To remain viable, an insurer needs to charge premiums at rates sufficient to cover expected benefit costs.
A setting like in Case 1 or Case 3—where the insurer knows the nature of the risk in advance and has high confidence in its estimate of expected benefit costs—does not exist in an open market. In an open market, the setting is usually similar to the one in Case 5. Each person who chooses to buy is underwritten. The insurer classifies the risk and then charges a price based on its expected benefit cost from that assessment.
For more about how insurers classify risks, or “risk classification,” see On Risk Classification,1 the Academy Risk Classification Work Group’s November 2011 monograph on the subject.
The less certain an insurer is about the expected cost of benefits, the higher the total risk margin the insurer is inclined to add to the premium. This margin can include a small-pool risk margin (described in Case 2) and a mix-of-sales risk margin (Case 4). Often, all consumers who buy from the insurer pay these risk margins, whether they individually contribute to the risk or not.
The insufficient-data risk margin (first described in Case 6) reflects uncertainty about the expected benefit cost of a risk due to an insufficient amount of credible data about that risk. For example, uncertainty about the amount of extra mortality risk linked to a certain high-risk activity, such as hang gliding, might cause insurers to add an insufficient-data risk margin to the premium rate only for a person expecting to take part in the activity—or instead (in some cases, at the person’s option) to exclude benefit coverage for death from such high-risk activity. This approach of impacting only those expecting to participate in high-risk activities allows insurers to keep their rates lower for consumers who do not expect to participate in such high-risk activities. For an example of a mandate to which this applies, see the section below on foreign travel.
Situations like the one in Case 4—where a mandate allows people of different ages to choose whether to buy but requires that the price be the same for everyone—are not equitable to consumers:
- Some consumers are high-cost risks, and they know it;
- Insurers cannot equitably classify and charge for (or exclude coverage for) these risks;
- Insurers must instead add a mix-of-sales risk margin to the premium rates for all consumers;
- The high-risk consumers select against the insurer because the cost is a bargain to them;
- Other consumers also know this and believe that their cost is not a good value to them;
- Many of these other consumers choose not to buy; and
- Insurers must worry about a rate spiral.
In these cases, the mandates can cause some consumers to pay a higher price for their coverage even though they are not higher-cost risks, and can cause other consumers to choose to go without coverage that they may need rather than pay the higher price. Also, because fewer policies are sold, the insurer has to consider the increased small-pool risk (Case 2), which could further increase the premiums paid by all consumers who buy coverage.
In some cases, an insurer might conclude that a particular mandate that limits insurers’ ability to equitably classify risks creates a market environment in which the insurer cannot be successful. In such cases, the insurer may choose not to participate in the market. This limits consumer choices in the market.
In the best cases, when an insurer expects little variance in benefit costs and little or no antiselection, then the total risk margin added to premium rates is lower and consumers pay a lower cost for the benefit.
Listed below are some mandates, enacted or considered, that limit insurers’ ability to equitably classify risks and charge for them, and what the results have been or may be.
HIV/AIDS—When the HIV/AIDS crisis emerged, life (and health) insurers would not insure people who had tested positive for, were being treated for, or showed symptoms of HIV/AIDS. Expected benefit costs were either too high or too uncertain. Some insurers considered HIV/AIDS testing for all people applying for policies. But such a practice would increase costs for all consumers who buy policies.
Policymakers wanted all people to have an “equal right” to buy life insurance without having to share information about their HIV/AIDS illness. Insurers maintained that the issue for them wasn’t privacy or unfair discrimination but rather cost. The expected cost of the antiselection risk was too high for all consumers to pay. A rate spiral would occur. This was illustrated above in Case 4 (albeit based on age, not medical condition) and in the fourth basic insurance principle. It is deemed no different than mandating the same rule for other people who know they have a serious medical condition, such as advanced-stage cancer.
At one point, policymakers in one jurisdiction (District of Columbia) did not permit insurers to use any test for HIV/AIDS. In response to this, several insurers unilaterally decided not to offer life insurance products there. This meant that all consumers in that jurisdiction had fewer product choices.
Eventually, testing for HIV/AIDS became allowed in all jurisdictions, although according to rules that varied by jurisdiction. But because an insurer cannot, in most states, use the results of an HIV/AIDS test to classify (or reject) coverage unless the insurer pays for its own test to be administered, insurers have to add the costs of the additional HIV/AIDS testing into their premium rates for all purchasers—the costs for all consumers who buy policies are increased.
Genetic testing—Genetic testing is an emerging science. Several types of tests have been created. Some of the testing companies claim to compare a person’s longevity relative to others. Although promising, insurers generally do not believe the science has yet matured to the point of meeting a rigorous standard of statistical credibility.
Insurers do not want to test every person choosing to buy life insurance. There are many tests—some costly—and their results currently may not be sufficiently credible. It would add costs for all consumers who buy policies, without demonstrated benefits. Insurers want to know only what each person already knows about the results of any genetic testing that has been done. This follows from our fourth basic insurance principle described above, as exemplified by Case 4. Permitting the insurer to know what the person already knows in classifying the risk reduces the antiselection risk described in Case 4 and the need to add extra risk margin to the premiums paid by all consumers.
This approach to genetic testing has a consumer benefit of avoiding the costs of testing and of antiselection on all policies.
But this approach has one admitted drawback for some consumers. Some might want to get tested now but are concerned that, later, the results of the test could affect their insurance premium rate (or even their insurability)—especially if the results are later deemed to be reliable. While this may be true, it is no different than a person considering getting screened now for diabetes or for cancer (the test results for which are already deemed reliable) and later applying for insurance.
Foreign travel—Insurers want to know if people intend to travel abroad in certain regions that may have heightened mortality risk. The information is needed to equitably classify the risk. This is no different than a person who intends to participate in a high-risk occupation or avocation (e.g., hang gliding). As shown in Case 4 and Case 5, the insurer needs to know what the person knows about the risk to prevent antiselection. Otherwise, a mix-of-sales risk margin would be needed and all consumers who buy would need to pay more to cover the higher expected benefit costs.
In addition, as in the hang gliding example, there might be insufficient data to credibly assess the additional mortality risk of travel in a certain region. Either an insufficient-data risk margin could be added to the premium rate (as shown in Case 6) or the risk could be excluded from coverage (in some cases, at the person’s option).
In some states, even though people know that they intend to travel in such areas and tell the insurer, a mandate prohibits the insurer from using that information in classifying the risk. And there is no limit on the face amount of coverage to which this mandate applies. As a result, all consumers who buy must pay for the extra risk.
Unisex life insurance premium rates—While stated as a limit on rates, this is really a limit on classifying risk. As in Case 4, where rates were not allowed to vary by age, the same issue arises here, where expected costs for males are higher than for females: The insurer needs to guess at a mix of sales by gender, and add a mix-of-sales risk margin to the gender-neutral (or “unisex”) premium rates. A rate spiral is likely, with most of the ultimate buyers being males. This means that many female consumers will choose not to buy, going without coverage they may need rather than buying insurance they view as a poor value to them.
Unisex lifetime-annuity premium or income rates—This is similar to the life insurance issue above, but in reverse. As in Case 8, the insurer has to guess how many males and how many females will decide to buy an annuity, in order to arrive at a uniform set of “equal” rates for males and females that will cover the total expected costs. This means that all males who choose to buy would have to pay a higher cost for their annuities than if rates could reflect gender. A rate spiral might discourage more and more males from purchasing the annuities that they may need, until only females purchase annuities. In an open market, the ideal situation is more like Case 9.Eventually, testing for HIV/AIDS became allowed in all jurisdictions, although according to rules that varied by jurisdiction. But because an insurer cannot, in most states, use the results of an HIV/AIDS test to classify (or reject) coverage unless the insurer pays for its own test to be administered, insurers have to add the costs of the additional HIV/AIDS testing into their premium rates for all purchasers—the costs for all consumers who buy policies are increased.
Genetic testing—Genetic testing is an emerging science. Several types of tests have been created. Some of the testing companies claim to compare a person’s longevity relative to others. Although promising, insurers generally do not believe the science has yet matured to the point of meeting a rigorous standard of statistical credibility.
Insurers do not want to test every person choosing to buy life insurance. There are many tests—some costly—and their results currently may not be sufficiently credible. It would add costs for all consumers who buy policies, without demonstrated benefits. Insurers want to know only what each person already knows about the results of any genetic testing that has been done. This follows from our fourth basic insurance principle described above, as exemplified by Case 4. Permitting the insurer to know what the person already knows in classifying the risk reduces the antiselection risk described in Case 4 and the need to add extra risk margin to the premiums paid by all consumers.
This approach to genetic testing has a consumer benefit of avoiding the costs of testing and of antiselection on all policies.
But this approach has one admitted drawback for some consumers. Some might want to get tested now but are concerned that, later, the results of the test could affect their insurance premium rate (or even their insurability)—especially if the results are later deemed to be reliable. While this may be true, it is no different than a person considering getting screened now for diabetes or for cancer (the test results for which are already deemed reliable) and later applying for insurance.
Foreign travel—Insurers want to know if people intend to travel abroad in certain regions that may have heightened mortality risk. The information is needed to equitably classify the risk. This is no different than a person who intends to participate in a high-risk occupation or avocation (e.g., hang gliding). As shown in Case 4 and Case 5, the insurer needs to know what the person knows about the risk to prevent antiselection. Otherwise, a mix-of-sales risk margin would be needed and all consumers who buy would need to pay more to cover the higher expected benefit costs.
In addition, as in the hang gliding example, there might be insufficient data to credibly assess the additional mortality risk of travel in a certain region. Either an insufficient-data risk margin could be added to the premium rate (as shown in Case 6) or the risk could be excluded from coverage (in some cases, at the person’s option).
In some states, even though people know that they intend to travel in such areas and tell the insurer, a mandate prohibits the insurer from using that information in classifying the risk. And there is no limit on the face amount of coverage to which this mandate applies. As a result, all consumers who buy must pay for the extra risk.
Unisex life insurance premium rates—While stated as a limit on rates, this is really a limit on classifying risk. As in Case 4, where rates were not allowed to vary by age, the same issue arises here, where expected costs for males are higher than for females: The insurer needs to guess at a mix of sales by gender, and add a mix-of-sales risk margin to the gender-neutral (or “unisex”) premium rates. A rate spiral is likely, with most of the ultimate buyers being males. This means that many female consumers will choose not to buy, going without coverage they may need rather than buying insurance they view as a poor value to them.
Note that many employer-sponsored pension plans are subject to a similar gender-neutral limit. But these plans are really more like Case 7, where there is a closed system and people do not choose whether to participate or how much to buy. It is not an open market, and it can be structured to be financially sound for that reason.
Mandates That Require Certain Rights or Benefits on New Policies
Like many other products offered in an open market, life insurance policies and annuity contracts can be designed with many “built-in” features and benefits. For example, some life policies include a built-in accidental death benefit, whereas others do not—just as some automobiles come with climate control as “standard equipment,” and, with others, it is offered as an extra-cost option.
Whether it is life insurance or new cars, a company designs a product with the combination of features it thinks customers will value the most. Why is this important? In an open market, consumers will decide to purchase based on their views about the value of the product to them.
When new (or increased) contractual rights or benefits for annuity or life insurance products are required by mandate, the expected extra costs of the mandate can be large or small. If they are perceived by an insurer to be significant, the insurer would generally have the following two options for recovering the extra cost of the benefit added to new policies:
- Increase the cost of the product, in which case—
- All consumers who buy the product will pay a higher cost for the product even though some, or many, may never use the new benefit;
- Some consumers might buy a lower amount of the product in order to keep their total costs from increasing, thereby receiving a lower amount of the benefit they need; and
- Some consumers who might need the product (but not the mandated benefit) may instead choose not to buy the product because they view the cost to be too high;
- Keep the cost of the product the same, possibly by reducing or dropping other benefits in the product, in which case all consumers will have such benefits reduced or eliminated.
Listed below are some examples of mandates that increase the cost of insurance for all consumers. The first example below is actually a case of a long-standing mandate becoming a significant cost issue for consumers due to an abrupt change in economic conditions.
Fixed policy-loan rates—Life insurance policies with cash values normally provide for loans. To set policy rates, insurers must estimate how much of the invested assets supporting policy values normally held in the company’s general investment account will be diverted to policy loans. If the loan interest rate is much lower than the rate earned on other investments, the impact can be large. When more is invested in policy loans at a relatively low rate, less can be invested elsewhere at a relatively higher rate. This impacts the overall investment returns that can be allocated among all policies to keep down costs.
Historically, state insurance laws have mandated that interest can be charged for policy loans at a fixed annual rate not to exceed some maximum, such as 6 or 8 percent. During the 1980s, rates of return available on personal investments spiked. Consumers began to realize that they could borrow from their policies at a relatively low rate of interest and invest that borrowed money in the market for an easy profit. In fact, some consumers were purchasing new policies just for that purpose. This is a form of interest-rate antiselection called “disintermediation.”
When insurers had to sell off long-term investments in order to obtain the liquid capital needed to provide these policy loans, the losses realized on those investments were either borne by investors (stock insurers) or passed along to policyowners in the form of lower policy dividends (mutual insurers), raising policy costs for all consumers.
To address this antiselection problem, insurers developed an “adjustable loan rate” provision. Until state laws were changed to allow insurers to use this new provision, all consumers who had policies with fixed loan rates were exposed to the risk of this antiselection.
‘Free-look’ provision—All life insurance policies and annuity contracts have a right-to-return, or “free-look,” provision. It provides the new contract owner a period of time after receiving the contract to review it and, if not satisfied with it, the right to return it for a full refund without penalty. State insurance laws generally require that this free-look period run for at least 10 days after taking delivery of the new contract.
The mandated right to return a newly entered contract—one involving a significant financial commitment and containing several provisions that might not be thoroughly discussed and understood at the time of application—is not unique to the life insurance industry. But unlike other types of contracts, the life insurance coverage during the free-look period is not “free.” These contracts do not delay commencement of coverage until completion of the free-look period.
Deaths do occur before the end of the free-look period, and death claims are paid for deaths during this period. Hence, there is a real cost for the life insurance coverage provided during this period. And the cost for this coverage under policies returned under the free-look period is borne by all the other contract owners—consumers of insurance coverage—who do not return their policies under this provision.
To extend the free-look period only increases the costs borne by the remaining consumers who purchased coverage. This issue can be exacerbated in cases where the insurer also is limited by another mandate that limits risk classification, such as not being able to consider the foreign-travel plans of applicants.
For example, with a mandated 30-day free-look period and such a mandated limit on risk classification, a person planning a trip to a potentially dangerous region might look at this combination as a way to obtain a complete medical examination (through the underwriting process) and a high amount of valuable “travel insurance” for free—paid for by the other consumers who purchase insurance coverage with the insurer and don’t return it. As with the fixed policy-loan rates, this can be a form of anti selection.
So, to extend the currently mandated free-look period could increase consumer costs—an unintended result—in several possible ways.
Surrender charges on flexible-premium contracts—State nonforfeiture laws mandate that owners of annuity and life insurance contracts who surrender their contracts do not forfeit the unused value of their accumulated premium payments. For fixed-premium life insurance contracts such as whole life, the mandate prescribes minimum cash surrender values calculated assuming recovery of insurers’ initial sales and administrative expenses is amortized over time.
Flexible-premium contracts such as universal life accumulate account values, based on actual premium amounts paid, that will vary from contract to contract. Deducting a surrender charge from the account value on surrender is a way to recover such unamortized initial expenses at that time. So nonforfeiture laws for flexible-premium contracts attempt to limit surrender charges on these contracts in a way intended to generate cash surrender values on a par with those for fixed-premium contracts.
To further limit surrender charges on flexible-premium contracts can be viewed by policymakers as increasing benefits to consumers who surrender their contracts for cash. While this might be true for surrendering contract owners, it would increase the losses that must be passed along to the remaining contract owners in higher costs. In effect, the remaining contract owners would be paying the cost of the additional cash benefits provided to the surrendering contract owners—an unintended result of such a mandate.
Minimum death benefits relative to premiums—Life insurance protection is needed by many types of consumers for many purposes. In some cases, it can be used by older consumers as a way to defray the costs of final expenses. The benefit amounts needed in these cases usually are relatively small. To meet the needs of the final-expenses market, classification (underwriting) requirements are often relaxed, which increases the expected average costs per thousand dollars of benefit. Because of this, the higher average age of buyers, and the fixed business expenses that must be covered by all policies, the premiums for these lower-amount policies relative to the benefits normally are higher than for most other policies.
Several years ago, policymakers in one jurisdiction mandated that, for most life insurance policies with such smaller benefit amounts, the death benefit cannot be lower than the premiums, accumulated at interest, during the first several years.2 This is a special case in this category of mandating benefits—it’s really mandating that premiums meet certain limits.
Similar to the result mentioned above in the HIV/AIDS section, several life insurers unilaterally decided not to offer lower-amount life insurance policies in this jurisdiction. And similar to the HIV/AIDS situation, this meant that all consumers in this jurisdiction needing lower amounts had fewer product choices.
This mandate has existed for many years. It originally applied to all life insurance policies with a minimum death benefit under $25,000 and was based on accumulating premiums at a fixed rate of 5 percent. In recent years, however, the mandate has been scaled back to affect only policies under $5,000 and accumulate premiums at a lower rate that is based on current market rates. Now, many more consumers in this market have more product choices.
Conclusion
In this issue brief, the following basic principles for annuity and life insurance benefits offered in the open market were discussed:
- To remain viable, insurers need to charge premiums at rates sufficient to cover expected benefit costs;
- If insurers are unsure of the level of expected benefit costs, they are likely to add extra risk margins to the rates in order to cover the risk of uncertainty, which increases consumer costs;
- Consumers are likely to decide to buy a product based on their view about the value of that product to them;
- If consumers can choose whether to buy the product and insurers are not permitted to use what consumers know about their own risks in setting rates, then consumers who believe they are high-cost risks are likely to select against the insurer; and
- If insurers can underwrite each risk and can equitably classify the risks, then insurers are more certain about the level of expected benefit costs and are less likely to add extra risk margins, which will keep consumer costs down.
When policymakers consider enacting mandates affecting annuity and life insurance products offered in the private insurance market, the possible impacts of such mandates on consumer cost and consumer choice can be:
- All consumers who purchase the affected products could pay a higher cost;
- Some consumers who know they are getting a bargain will pay a lower cost than they should pay;
- Other consumers will pay a higher cost than they should pay, subsidizing the consumers getting the bargain;
- Some consumers will purchase less of the benefit than they need, in order to keep their costs from increasing;
- Some consumers will choose not to purchase the benefit that they need because they deem the cost to be too high; and
- In some cases, consumers will be left with fewer product choices because insurers will choose not to participate in the market.
| 1. | Available online at www.actuary.org/files/publications/RCWG_RiskMonograph_Nov2011.pdf. |
| 2. | State of Washington, WAC 284-23-550 |
Share
